
# import all packages and set plots to be embedded inline
from requests import get
from zipfile import ZipFile
from io import StringIO, BytesIO
import numpy as np
import pandas as pd
import missingno as ms
import matplotlib.pyplot as plt
import seaborn as sb
import geopandas as gpd
import folium
from shapely.geometry import Point, Polygon
import haversine as hs


%matplotlib inline


#Download the fordgobike data from the Udacity provided link
url = 'https://video.udacity-data.com/topher/2020/October/5f91cf38_201902-fordgobike-tripdata/201902-fordgobike-tripdata.csv'
data_csv = get(url)
data_csv
#<Response [200]> = Success

<Response [200]>


#Store the downloaded data in a csv file and verify it
df = pd.read_csv(StringIO(data_csv.content.decode('utf-8')))
df.head()


#Check the size of the data (rows,columns)
df.shape

(183412, 16)


#Investigate the column info to see are there null values and incorrect data types. 
df.info(show_counts = True)

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 183412 entries, 0 to 183411
Data columns (total 16 columns):
 #   Column                   Non-Null Count   Dtype  
---  ------                   --------------   -----  
 0   duration_sec             183412 non-null  int64  
 1   start_time               183412 non-null  object 
 2   end_time                 183412 non-null  object 
 3   start_station_id         183215 non-null  float64
 4   start_station_name       183215 non-null  object 
 5   start_station_latitude   183412 non-null  float64
 6   start_station_longitude  183412 non-null  float64
 7   end_station_id           183215 non-null  float64
 8   end_station_name         183215 non-null  object 
 9   end_station_latitude     183412 non-null  float64
 10  end_station_longitude    183412 non-null  float64
 11  bike_id                  183412 non-null  int64  
 12  user_type                183412 non-null  object 
 13  member_birth_year        175147 non-null  float64
 14  member_gender            175147 non-null  object 
 15  bike_share_for_all_trip  183412 non-null  object 
dtypes: float64(7), int64(2), object(7)
memory usage: 22.4+ MB


#Show the sum of null entries for each column
df.isna().sum()

duration_sec                  0
start_time                    0
end_time                      0
start_station_id            197
start_station_name          197
start_station_latitude        0
start_station_longitude       0
end_station_id              197
end_station_name            197
end_station_latitude          0
end_station_longitude         0
bike_id                       0
user_type                     0
member_birth_year          8265
member_gender              8265
bike_share_for_all_trip       0
dtype: int64


# describe the data to generate descriptive statistics
df.describe()


#Verify if the data contains duplicate rows
df.duplicated().sum()

0


#Visualize missing data with missingno
ms.matrix(df);


#Remove missing values.
df.dropna(subset = ["start_station_id"], inplace = True)
#Replace NULL values with 0
df.member_birth_year.fillna(0, inplace = True)
#Replace NULL values with 'not defined'
df.member_gender.fillna("not defined", inplace = True)


#Convert column to the correct dtypes
df['start_time'] = pd.to_datetime(df['start_time'], format = "%Y-%m-%d ")
df['end_time'] = pd.to_datetime(df['end_time'], format = "%Y-%m-%d ")
df['duration_sec'] = df['duration_sec'].astype(int)
df['start_station_id'] = df['start_station_id'].astype(str)
df['end_station_id'] = df['end_station_id'].astype(str)
df['start_station_latitude'] = df['start_station_latitude'].astype(float)
df['start_station_longitude'] = df['start_station_longitude'].astype(float)
df['end_station_latitude'] = df['end_station_latitude'].astype(float)
df['end_station_longitude'] = df['end_station_longitude'].astype(float)
df['start_time'] = pd.to_datetime(df['start_time'], format = "%Y-%m-%d ")
df['bike_id'] = df['bike_id'].astype(str)
df['member_birth_year'] = df['member_birth_year'].astype(int)


#Verify data cleaning
ms.matrix(df);


#Make a new dataframe for geocode data analysis
geo_df = pd.DataFrame()
geo_df['start_lat'] = df['start_station_latitude']
geo_df['start_long'] = df['start_station_longitude']
geo_df['end_lat'] = df['end_station_latitude']
geo_df['end_long'] = df['end_station_longitude']
geo_df['duration_sec'] = df['duration_sec']
geo_df['date'] = df['start_time']
geo_df['member_birth_year'] =  df['member_birth_year']
geo_df['member_gender'] = df['member_gender']
geo_df = geo_df.astype(str)
i=0
rows = geo_df.shape[0]

geo_df.dtypes
geo_df['trips'] = "Start Location: " + ',' + geo_df['start_lat'].map(str) + ',' + geo_df['start_long'].map(str) + ',' + "End Location: " + ',' + geo_df['end_lat'].map(str) + ',' + geo_df['end_long']
#trip_counts shows counts of the most popular journeys 
trip_counts = geo_df['trips'].value_counts()
print(trip_counts.head(5))

Start Location: ,37.77588,-122.39317,End Location: ,37.795392,-122.394203                        337
Start Location: ,37.795392,-122.394203,End Location: ,37.80477,-122.403234                       314
Start Location: ,37.80889393398715,-122.25646018981932,End Location: ,37.8090126,-122.2682473    310
Start Location: ,37.80477,-122.403234,End Location: ,37.79413,-122.39443                         285
Start Location: ,37.8090126,-122.2682473,End Location: ,37.80889393398715,-122.25646018981932    284
Name: trips, dtype: int64


top_trips = trip_counts.head(10)
top_trips.head(5)

Start Location: ,37.77588,-122.39317,End Location: ,37.795392,-122.394203                        337
Start Location: ,37.795392,-122.394203,End Location: ,37.80477,-122.403234                       314
Start Location: ,37.80889393398715,-122.25646018981932,End Location: ,37.8090126,-122.2682473    310
Start Location: ,37.80477,-122.403234,End Location: ,37.79413,-122.39443                         285
Start Location: ,37.8090126,-122.2682473,End Location: ,37.80889393398715,-122.25646018981932    284
Name: trips, dtype: int64


#Show most popular journeys as a percentage
trip_counts_percentage = geo_df['trips'].value_counts(normalize=True)
print("Key : {} , Value : {}".format(trip_counts_percentage.index[0],trip_counts_percentage[0]))

Key : Start Location: ,37.77588,-122.39317,End Location: ,37.795392,-122.394203 , Value : 0.0018393690472941625


print(trip_counts.index[0])

Start Location: ,37.77588,-122.39317,End Location: ,37.795392,-122.394203


print(trip_counts.values)

[337 314 310 ...   1   1   1]


#Get some metrics from the trip_counts data
print(len(trip_counts))
print(trip_counts.median())
print(trip_counts.min())
print(trip_counts.max())

23648
3.0
1
337


#Pie Chart to visualize the top journeys
values = top_trips.values
labels = top_trips.index
explode = (0.2, 0, 0, 0, 0, 0, 0, 0, 0, 0)
plt.pie(values, labels= values,explode=explode,counterclock=False, shadow=True)
plt.title('Top 10 Bike Journeys')
plt.legend(labels, loc='center left', bbox_to_anchor=(1.5, 0.5))
plt.show()


#top 500 is the most the jupyter notebook can handle without crashing or becoming to slow.
n = 500
top_trips = geo_df['trips'].value_counts().index.tolist()[:n]
#print(top_trips)


print(geo_df.shape[0])

183215


#Use Folium to plot the journeys on a map
#Green markers are start points, red are finish points
#Hover the marker to get the journey number
my_map = folium.Map(location=(37.7693053,-122.4268256), zoom_start=11);
i=0
limit = len(top_trips)
while i < limit:
    trip = top_trips[i].split(',')
    #print(trip)
    #print('Trip: ',i,'Start LatLong: ',trip[0],trip[1],'End LatLong: ',trip[2],trip[3])
    folium.Marker(location=(trip[1],trip[2]),popup='Start trip:'+str(i),icon=folium.Icon(color='green',icon='circle')).add_to(my_map)
    folium.Marker(location=(trip[4],trip[5]),popup='End trip:'+str(i),icon=folium.Icon(color='red', icon='square')).add_to(my_map)
    i+=1

display(my_map)


#Use Seaborn to plot the jouney in a scatterplot
axes, figure = plt.subplots(figsize = (10,5))
sb.scatterplot(data = df[df.start_station_id.isnull()], x = "end_station_longitude", y = "end_station_latitude", alpha = 0.15, s = 200)
sb.scatterplot(data = df.dropna(subset=["end_station_id"]).sample(50000), x = "start_station_longitude", y = "start_station_latitude", alpha = 0.15, s = 200)
plt.xlim(-121.8,-122.5)
plt.ylim(37.2,38.2)
plt.xlabel("Latitude");
plt.ylabel("Longitude");
plt.tight_layout()


#https://stackoverflow.com/questions/29545704/fast-haversine-approximation-python-pandas
#this function teakes start Lat Long and Dest Lat long and caculates the distance between them
from math import radians, cos, sin, asin, sqrt
from time import time
#start time
t0 = time()
print('This calculation is slow (minutes), please be patient . . .')
def haversine(lon1, lat1, lon2, lat2):
    """
    Calculate the great circle distance between two points 
    on the earth (specified in decimal degrees)
    """
    # convert decimal degrees to radians 
    lon1, lat1, lon2, lat2 = map(radians, [lon1, lat1, lon2, lat2])
    # haversine formula 
    dlon = lon2 - lon1 
    dlat = lat2 - lat1 
    a = sin(dlat/2)**2 + cos(lat1) * cos(lat2) * sin(dlon/2)**2
    c = 2 * asin(sqrt(a)) 
    km = 6367 * c
    return round(km,2) 

i=0
for index, row in df.iterrows():
    trip = geo_df['trips'].iloc[i].split(',')
    #print(trip[1],trip[2], trip[4], trip[5])
    geo_df.loc[index, 'distances'] = haversine(float(trip[1]), float(trip[2]), float(trip[4]), float(trip[5]))
    i+=1

#End time
print("Time taken: {} seconds\nfor {} observations".format(time()-t0, len(geo_df)))

This calculation is slow (minutes), please be patient . . .
Time taken: 180.83300828933716 seconds
for 183215 observations


#verify the distance calculations
print(geo_df['distances'].head(5))

0    0.36
1    0.96
2    2.64
3    0.27
4    2.65
Name: distances, dtype: float64


#Replace 0 values with NaN, so as not to be counted
geo_df['distances'].replace(0, np.nan, inplace=True)
distance_counts = geo_df['distances'].value_counts(dropna=True)
#Add the distances column to the main dataframe
df['distances'] = geo_df['distances']


#function to plot a piechart
def pie_Chart(values, labels, title, explode):
    top_dist = distance_counts.head(10)
    explode = explode
    plt.pie(values, labels= values,explode=explode,counterclock=False, shadow=True)
    plt.title(title)
    plt.legend(labels, loc='center left', bbox_to_anchor=(1.5, 0.5))
    plt.show()


#plot piechart of Most counted Distances in km using function
top_dist = distance_counts.head(10)
values = top_dist.values
labels = top_dist.index
title = 'Most counted Distances in km'
explode = (0.3, 0, 0, 0, 0, 0, 0, 0, 0, 0)

pie_Chart(values, labels, title, explode)


#plot piechart of Longest Distances in km using function
dist = geo_df['distances'].value_counts()
sorted_dist = dist.sort_index(ascending=False)
top_dist = sorted_dist.head(10)
indexs = top_dist.index
values = top_dist.values
title = 'Longest Distances in km'
explode = (0.3, 0, 0, 0, 0, 0, 0, 0, 0, 0)
pie_Chart(indexs, values, title, explode)


#Convert columns to numerical and date dtypes, so that I can preform mathematical functions on values.
df['distances'] = df['distances'].astype(float)
df['duration_sec'] = df['duration_sec'].astype(int)
df['member_birth_year'] = df['member_birth_year'].astype(float)
df['date'] = pd.to_datetime(geo_df['date'], format = "%Y-%m-%d ")
df.dtypes

duration_sec                        int64
start_time                 datetime64[ns]
end_time                   datetime64[ns]
start_station_id                   object
start_station_name                 object
start_station_latitude            float64
start_station_longitude           float64
end_station_id                     object
end_station_name                   object
end_station_latitude              float64
end_station_longitude             float64
bike_id                            object
user_type                          object
member_birth_year                 float64
member_gender                      object
bike_share_for_all_trip            object
distances                         float64
date                       datetime64[ns]
dtype: object


#What is the longest distance cycled?
df['distances'].max()

63.76


#Calculate age of users by (data_year - birth_year) and add to the main dataframe
df["age"] = df["member_birth_year"].apply(lambda x: 2019 - int(x))
geo_df["age"] = df["age"]


#cut month_year, store in geo_df, as data isn't neeeded in main data frame
geo_df['month_year'] = pd.to_datetime(df["start_time"]).dt.to_period('M')


#cut day_month_year. 
geo_df['day_month_year'] = pd.to_datetime(df["start_time"]).dt.to_period('D')


#calculate day of the week from datetime
geo_df["dayofweek"] = df["start_time"].apply(lambda x: x.dayofweek)


#calculate start and end hour of journey.
geo_df["start_hr"] = df["start_time"].apply(lambda x: x.hour)
geo_df["end_hr"] = df["end_time"].apply(lambda x: x.hour)


#Create Decade age bins
bins = [x for x in range(10,101, 10)]
df["age_bins"] = pd.cut(df.age, bins = bins, precision = 0, include_lowest=False)


#verify data
df[["age", "age_bins"]].head()


#verify data
geo_df[["month_year", "day_month_year","dayofweek","start_hr","end_hr"]].head()


# Plotting time vs. Distances

plt.figure(figsize=(17, 8))

plt.xlabel('Date')
plt.ylabel('Distances (km)')

plt.plot(df['date'],df["distances"])
plt.title('Distances Cycled over time');


#Plot the largest journey to see was it feasible
my_map = folium.Map(location=(37.7693053,-122.4268256), zoom_start=8);
folium.Marker(location=(37.7896254,-122.400811 ),popup='Start of 69.47km trip:',icon=folium.Icon(color='green',icon='circle')).add_to(my_map)
folium.Marker(location=(37.3172979,-121.884995),popup='End of 69.47km trip:',icon=folium.Icon(color='red', icon='square')).add_to(my_map)
display(my_map)


#View distances max, min & mean
print(df['distances'].max())
print(df['distances'].min())
print(df['distances'].mean())

63.76
0.02
1.5119350253354733


#plot count of bike journey durations
#use logarithmic scale
bin_edges = 10 ** np.arange(0, 5, 0.1)
ticks = [30,100,300,1000,3000,10000,30000,100000]
fig, axes = plt.subplots(figsize = (12,5), dpi = 110)
labels = ['{}'.format(v) for v in ticks]
plt.hist(data = df, x ='duration_sec', bins = bin_edges);
plt.xscale("log");
plt.xlim(30, 10000);
plt.xticks(ticks,labels);
plt.title("Count of duration of bike journey");
plt.xlabel("Duration of journey in Seconds");
plt.ylabel("Count of durations");


#plot count of birth years of members
bin_edges = np.arange(df['member_birth_year'].min(),df['member_birth_year'].max(), 1)
fig, axes = plt.subplots(figsize = (12,5), dpi = 110)
plt.hist(data = df, x ='member_birth_year', bins = bin_edges);
plt.xlim(1930, 2005);
plt.title("Count of Birth year of members");
plt.xlabel("Year of birth of members");
plt.ylabel("Count of year of births");


#plot count of user types Subscribers Vs Customers
values = df.user_type.value_counts()

fig, ax = plt.subplots(figsize = (10,5), dpi = 80)
sb.countplot(x = "user_type", data = df, order=values.index, palette = "cividis");

plt.title("Users By Type");
plt.xlabel("User Type");
plt.ylabel("Count of user type for each journey");


#count trips by gender
fig, ax = plt.subplots(figsize = (10,5), dpi = 80)
sb.countplot(x = "member_gender", data = df,  order=df.member_gender.value_counts().index, palette = "cividis");

plt.title("Users By Gender");
plt.xlabel("Gender of member");
plt.ylabel("Count of genders for each trip");


#plot piechart of Most counted Distances in km using function

gender = df['member_gender'].value_counts()
sorted_gender = gender.sort_values(ascending=False)
values = sorted_gender.values
indexs = sorted_gender.index
title = 'Percentages of Gender that used the service'
explode = (0.3, 0, 0, 0)

pie_Chart(values, labels, title, explode)


#plot histogram to show count of Individual journey distances
#Show the 1st km in 10 divisions for detail
fig, ax = plt.subplots(figsize = (12,5), dpi = 80)
bin_size = 0.1
bin_edges = np.arange(0,geo_df.distances.max()+bin_size,bin_size)
ticks = [0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1,1.2,1.4,1.6,1.8,2,2.5,3,4,5,6,7,8,9,10,12,14,16]
labels = ['{}'.format(v) for v in ticks]
plt.hist(data = geo_df, x ='distances', bins = bin_edges);
plt.xscale("symlog");
plt.xlim(0.1, 16);
plt.xticks(ticks,labels);
plt.title("Count of Individual journey distances");
plt.xlabel("Distance in Kilometres");
plt.ylabel("Count of trip distances");


#plot histogram to show count of ages
fig, ax = plt.subplots(figsize = (15,5), dpi = 100)
color = sb.color_palette("cividis_r")
sb.countplot(x = "age", data = df.query("age < 80").sort_values("age"));
plt.title("Count of Bike user ages");


#Show counts of rides by day of the week
#It can be seen that Thursday is peak and the weekend is quieter
fig, ax = plt.subplots(figsize = (16,5)) 
sb.countplot(x = "dayofweek", data = geo_df, palette = "cividis_r");
plt.title("Count of rides by Day of the Week");
days = ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday"]
plt.xticks(range(len(days)), days, size='medium');


#convert distances to float for mathematical calculations
df['distances'] = df['distances'].astype(float)
df.dtypes

duration_sec                        int64
start_time                 datetime64[ns]
end_time                   datetime64[ns]
start_station_id                   object
start_station_name                 object
start_station_latitude            float64
start_station_longitude           float64
end_station_id                     object
end_station_name                   object
end_station_latitude              float64
end_station_longitude             float64
bike_id                            object
user_type                          object
member_birth_year                 float64
member_gender                      object
bike_share_for_all_trip            object
distances                         float64
date                       datetime64[ns]
age                                 int64
age_bins                         category
dtype: object


#show gender count vs distances in violin plot
sb.violinplot(data = df.query("distances <= 6"),
               x = 'member_gender',y='distances',
               color = sb.color_palette()[0],inner='quartile');
plt.xticks(rotation=1);
plt.title("Common distances cycled by Gender");


#show gender count vs distances in violin plot
sb.violinplot(data = df.query("duration_sec <= 4000"),
               x = 'member_gender',y='duration_sec',
               color = sb.color_palette()[0],inner='quartile');
plt.xticks(rotation=1);
plt.title("Common Durations Cycled by Gender");


#show user type vs birth year in violin plot
sb.violinplot(data = df.query("member_birth_year >= 1940"),
               x = 'user_type',y='member_birth_year',
               color = sb.color_palette()[0],inner='quartile');
plt.xticks(rotation=1);
plt.title("Common birth years by user type");


#show user type vs duration in violin plot
sb.violinplot(data = df.query("duration_sec <= 4039.5"),
               x = 'user_type',y='duration_sec',
               color = sb.color_palette()[0],inner='quartile');
plt.title("Trip duration vs Customer type");
plt.xlabel("Customer type");
plt.ylabel("Trip duration");


sb.violinplot(data = df.query("duration_sec <= 4039.5"),x = 'user_type',y='distances',color = sb.color_palette()[0],inner='quartile');
plt.title("Trip duration vs Customer type");
plt.xlabel("Customer type");
plt.ylabel("Trip duration");


#plot 2d histogram to show count of births vs duration cycled
xbin = np.arange(df['member_birth_year'].min(), df['member_birth_year'].max()+5, 5)
ybin = np.arange(0, 5000, 200)
plt.hist2d(data = df,x = 'member_birth_year',y='duration_sec',cmin=0.5,cmap = 'viridis_r',bins=[xbin, ybin]);
plt.xlim(1940, 2005);
plt.colorbar();
plt.colorbar().set_label('Counts', rotation=270);
plt.title("Count of common birth years and durations cycled");
plt.xlabel("Birth Year");
plt.ylabel("Duration in Seconds");


#set plot dimensions
plt.figure(figsize=[14,6])

#create  a decade variable, using floor division (lowest integer divisor)
df['member_birth_decade'] = ((df['member_birth_year']//10)*10).astype(int)
                                                                     
#use decade
sb.stripplot(data = df.query('member_birth_decade>0'),
             x = 'member_birth_decade', y = 'distances', hue ='member_gender',
             jitter = 0.35, dodge = True, palette = 'Dark2', size = 3);

plt.legend(['Male','Female','Other'])
plt.title("Count of Gender,birth,year distances cycled");
plt.ylim((0,20));


#set plot dimensions
plt.figure(figsize=[14,6])

#create  a decade variable, using floor division (lowest integer divisor)
df['member_birth_decade'] = ((df['member_birth_year']//10)*10).astype(int)
                                                                     
#use decade
sb.stripplot(data = df.query('member_birth_decade>0'),
             x = 'member_birth_decade', y = 'duration_sec', hue ='member_gender',
             jitter = 0.35, dodge = True, palette = 'Dark2', size = 3);

plt.legend(['Male','Female','Other'])
plt.title("Count of Gender,birth,year distances cycled");
plt.ylim((0,90000));


#plot heatmap with seaborn counts of gender, user type and average distance cycled
cat_means = df.groupby(['member_gender', 'user_type']).mean()['distances']
cat_means = cat_means.reset_index(name = 'distance_avg')
cat_means = cat_means.pivot(index = 'user_type', columns = 'member_gender',
                            values = 'distance_avg')
sb.heatmap(cat_means, annot = True, fmt = '.3f',
           #set heat map to distance_avg mean
           cbar_kws = {'label' : 'mean(distance_avg)'});
plt.title("Count of user type,birth year Average distance mean");


#plot heatmap with seaborn counts of gender, user type and average duration cycled
cat_means = df.groupby(['member_gender', 'user_type']).mean()['duration_sec']
cat_means = cat_means.reset_index(name = 'duration_sec_avg')
cat_means = cat_means.pivot(index = 'user_type', columns = 'member_gender',
                            values = 'duration_sec_avg')
plt.title("Count of user type,gender Average distance mean");
sb.heatmap(cat_means, annot = True, fmt = '.3f',
           #set heat map to duration_avg mean
           cbar_kws = {'label' : 'mean(duration_sec_avg)'});


#plot 2D histogram showing counts of duration in Seconds vs Year of birth vs Customer type
xbin = np.arange(1940, df['member_birth_year'].max()+5, 5)
ybin = np.arange(0, 4500, 500)
#set grid size
grid = sb.FacetGrid(data = df, col = 'user_type',height=5)
grid.map(plt.hist2d, 'member_birth_year','duration_sec',cmin=0.5,cmap = 'cividis_r',bins=[xbin, ybin]);
plt.colorbar().set_label('Counts');

plt.subplots_adjust(top=0.9)
grid.fig.suptitle("Duration in Seconds vs Year of birth vs Customer type");

grid.set_ylabels("Duration");
grid.set_xlabels("Year of birth");


#plot 2D histogram showing counts of distances vs Year of birth vs Customer type
xbin = np.arange(1940, df['member_birth_year'].max()+5, 5)
ybin = np.arange(0, 20, 2)
#set grid size
grid = sb.FacetGrid(data = df, col = 'user_type',height=5)
grid.map(plt.hist2d, 'member_birth_year','distances',cmin=0.5,cmap = 'cividis_r',bins=[xbin, ybin]);
plt.colorbar().set_label('Counts');

plt.subplots_adjust(top=0.9)
grid.fig.suptitle("Distances vs Year of birth vs Customer type");

grid.set_ylabels("distances");
grid.set_xlabels("Year of birth");


df.to_csv("master_data_frame.csv", sep=',', encoding='utf-8')

	duration_sec	start_time	end_time	start_station_id	start_station_name	start_station_latitude	start_station_longitude	end_station_id	end_station_name	end_station_latitude	end_station_longitude	bike_id	user_type	member_birth_year	member_gender	bike_share_for_all_trip
0	52185	2019-02-28 17:32:10.1450	2019-03-01 08:01:55.9750	21.0	Montgomery St BART Station (Market St at 2nd St)	37.789625	-122.400811	13.0	Commercial St at Montgomery St	37.794231	-122.402923	4902	Customer	1984.0	Male	No
1	42521	2019-02-28 18:53:21.7890	2019-03-01 06:42:03.0560	23.0	The Embarcadero at Steuart St	37.791464	-122.391034	81.0	Berry St at 4th St	37.775880	-122.393170	2535	Customer	NaN	NaN	No
2	61854	2019-02-28 12:13:13.2180	2019-03-01 05:24:08.1460	86.0	Market St at Dolores St	37.769305	-122.426826	3.0	Powell St BART Station (Market St at 4th St)	37.786375	-122.404904	5905	Customer	1972.0	Male	No
3	36490	2019-02-28 17:54:26.0100	2019-03-01 04:02:36.8420	375.0	Grove St at Masonic Ave	37.774836	-122.446546	70.0	Central Ave at Fell St	37.773311	-122.444293	6638	Subscriber	1989.0	Other	No
4	1585	2019-02-28 23:54:18.5490	2019-03-01 00:20:44.0740	7.0	Frank H Ogawa Plaza	37.804562	-122.271738	222.0	10th Ave at E 15th St	37.792714	-122.248780	4898	Subscriber	1974.0	Male	Yes

	duration_sec	start_station_id	start_station_latitude	start_station_longitude	end_station_id	end_station_latitude	end_station_longitude	bike_id	member_birth_year
count	183412.000000	183215.000000	183412.000000	183412.000000	183215.000000	183412.000000	183412.000000	183412.000000	175147.000000
mean	726.078435	138.590427	37.771223	-122.352664	136.249123	37.771427	-122.352250	4472.906375	1984.806437
std	1794.389780	111.778864	0.099581	0.117097	111.515131	0.099490	0.116673	1664.383394	10.116689
min	61.000000	3.000000	37.317298	-122.453704	3.000000	37.317298	-122.453704	11.000000	1878.000000
25%	325.000000	47.000000	37.770083	-122.412408	44.000000	37.770407	-122.411726	3777.000000	1980.000000
50%	514.000000	104.000000	37.780760	-122.398285	100.000000	37.781010	-122.398279	4958.000000	1987.000000
75%	796.000000	239.000000	37.797280	-122.286533	235.000000	37.797320	-122.288045	5502.000000	1992.000000
max	85444.000000	398.000000	37.880222	-121.874119	398.000000	37.880222	-121.874119	6645.000000	2001.000000

Part I - Ford GoBike Dataset Exploration¶

by Darragh Merrick¶

Introduction¶

Dataset Overview and Notes¶

Example Topics/Questions¶

Preliminary Wrangling¶

Data Cleaning¶

Missing data¶

What is the structure of your dataset?¶

Exploratory Data Visualizations¶

Formula to calculate distance between 2 locations¶

Univariate Exploration¶

What is/are the main feature(s) of interest in your dataset?¶

What features in the dataset do you think will help support your investigation into your feature(s) of interest?¶

Discuss the distribution(s) of your variable(s) of interest. Were there any unusual points? Did you need to perform any transformations?¶

Of the features you investigated, were there any unusual distributions? Did you perform any operations on the data to tidy, adjust, or change the form of the data? If so, why did you do this?¶

Bivariate Exploration¶

Talk about some of the relationships you observed in this part of the investigation. How did the feature(s) of interest vary with other features in the dataset?¶

Did you observe any interesting relationships between the other features (not the main feature(s) of interest)?¶

Multivariate Exploration¶

Talk about some of the relationships you observed in this part of the investigation. Were there features that strengthened each other in terms of looking at your feature(s) of interest?¶

Were there any interesting or surprising interactions between features?¶

Conclusions¶

	age	age_bins
0	35	(30.0, 40.0]
1	2019	NaN
2	47	(40.0, 50.0]
3	30	(20.0, 30.0]
4	45	(40.0, 50.0]

	month_year	day_month_year	dayofweek	start_hr	end_hr
0	2019-02	2019-02-28	3	17	8
1	2019-02	2019-02-28	3	18	6
2	2019-02	2019-02-28	3	12	5
3	2019-02	2019-02-28	3	17	4
4	2019-02	2019-02-28	3	23	0